Disrupt Yourself or be Uberized
As I revisit some of my most recent conversations on digital disruption with automotive and logistics CIOs, it has become apparent to me that they are all working on proposals on how to change their organizations’ business models. Design, test, sell and service a car every 5,000 miles is no longer good enough and orchestrating bespoke transportation services is neither.
Per Robert Sanchez, CEO of Ryder (speaker at speaker at InformaticaWorld 2017 in San Francisco this month), these secular trends drive logistics outsourcing: Driver shortages, infringing minimum wage laws, shorter supply chains, the growth of eCommerce, pervasive emission and safety regulation and new engine technology (EV) and related maintenance cost.
Uber has recognized this with its acquisition of Otto. This puts pressure on third party logistics providers to reinvent on what basis they can compete. Uber providing local, commercial delivery services does the same. Mercedes, Volvo, Ford and GM are all working feverishly on accelerating their telematics work into autonomous vehicle development. Like Uber, they are running prototypes in cities like San Diego, Pittsburgh and so on.
McKinsey postulated a while back that fleet management, supply chain and inventory management processes specifically in the logistics space offer the largest potential for IoT (telematics).
What does this mean? Most experts agree that autonomous capabilities will see operational use on public streets via commercial trucking vehicles way before consumer vehicles. They will start with port to outside-of-town distribution center deliveries first, then making their way into suburban and lastly, more complex and denser metropolitan areas. Once this has created a commercially viable and safe track record, only then, in my estimation, will passenger vehicles receive the all-in approval to operate fully autonomously. Prior to this, autonomous function will be a location or speed-dependent, turn on/off function, just like your radar-enabled cruise control and governed in a supervised or at least semi-supervised fashion when it comes to the use of learning algorithms.
So why embark on this lengthy, costly, complex and risky endeavor?
Because the last decade has shown through the likes of Amazon, Airbnb and Uber that only the top 2 disrupters can collect 80% of the respective market’s profits (Bain&Co).
Apart from this opportunity, there is risk for the afore mentioned large incumbents in the OEM and 3PL (third party logistics) space, specifically around how to use the volume of data; Uber is so easily acquiring, managing and leveraging. Forbes Magazine labeled Uber a Big Data Company a while back. This has yet to happen for one of the big guys like the Detroit 3, Japanese or European OEMs or the likes of Hertz, Avis, DB Schenker, CSX, JB Hunt, etc.
Data standards, key players and their consortia scope, regulations and most profitable use cases are still emerging. The key telematics data use cases being worked on today and soon for autonomous driving include:
- Resource & Energy Management (think fuel contracts)
- Equipment & Employee Monitoring (think A/C failure and drive time)
- Human & Merchandise Health Monitoring (think refrigerated products and wearables)
- Physical Security (think trailer and warehouse access)
- Route Tracking & Optimization (think pickup scheduling based on vehicle/driver performance)
- Usage Based Insurance (think trip/voyage based premiums considering vehicle/driver performance)
- Warehouse Space Management (think RFID based industrial robot routing)
- Predictive Maintenance (think maintenance interval adjustment based on past and planned usage)
- Asset Sharing (think kick-backs to bespoke customers for using their empty trucks for commercial and consumer use)
One key angle here is for OEMs and 3PLs to embed themselves within their customers in a cooperative fashion. They need to understand the commercial challenges they experience to formulate meaningful technical solutions and commercial propositions.
This means that if your key commercial clients are electronics, general machinery and oil and gas exploration firms, understand their cost and revenue model, regulatory framework, key processes and what it means if a certain process fails or takes longer than necessary.
If the OEM or logistics firm can ensure fewer transaction (expedited shipping, spoilage, product returns, etc.) fails or require fewer resources to complete, they identified a potential improvement area.
Now, what internal insight is available and in what form (raw data, processed data capturing only outlier data points, analytics or actions like alerts or bids) can be served to this client to achieve the new outcome? What Uber-esque technical capabilities are missing to get there?
Also, given the impact to the client, what is this capability worth and how should it be charged; a subscription, a per-transaction fee or percentage, a flat upfront fee, a usage based percentage or a combination thereof? What can the client’s cash flow stomach?
I am a big proponent of testing use cases via smaller proof-of-concepts and once proven to be quickly rolled-out more widely.
What is your take on digital disruption? What is the most innovative and successful disruption you have seen using the power of data?